next up previous contents
Next: The Fan Observatory Bench Up: manual Previous: The GenI CCD Camera


The GenII CCD Camera (Spectroscopy)


(Rev. August 17, 2011)

System Description

The GenII CCD system uses the Generation II controller developed by San Diego State University (Leach group) and now managed under the company name of Astronomical Research Cameras, Inc. (ARC). The detector is an SI424A scientific grade CCD imager manufactured by Scientific Imaging Technologies, Inc. (SITe). This camera system is intended primarily use with the Fan Mountain Observatory Bench Spectrograph (FMOBS), which is described in detail elsewhere.

The SDSU controller is mounted directly on a CCD liquid nitrogen dewar with a serial fiber optic communications link leading from the controller housing to a PCI interface card in a Sun Ultra5 computer. For operation with the GenII controller, the toggle switch on the back panel of the computer should be set to ``GEN II''.

The power supply for the controller is a gray metal box with a switch which must be turned on for the controller to operate. Other hardware details of the system will depend on the spectrograph setup.

The Sun Ultra5 control computer is named crux and uses the Sun Solaris2.8 operating system with the CDE window system. The user interface to the CCD controller is a program called Voodoo developed by SDSU and modified locally for use at Fan Mountain. Images are stored on disk in FITS (*.fits) format and can be transferred from disk to magnetic tape (4mm DAT DDS4). A typical full frame requires 8.6MB of storage. Images can be displayed and analyzed using IRAF tasks. This manual does not explain the details of using IRAF, but IRAF has an extensive built-in help facility, and full documentation is available on the web at the IRAF Project Home Page.

CCD Camera Specifications

Table 1 summarizes the current configuration of the CCD camera, chip, and dewar.

Table 1: Current configuration of the CCD chip and camera.
Spec Value
Dewar IR Labs
  liquid nitrogen cooled, 1 ${\ell}$ capacity
  heating resistor
  hold time $\sim36$ hours
Chip SITe 2048 $\times$ 2048 CCD Imager
  back-illuminated, thinned to enhance blue response
Operating Temp. unstable above $-100^\circ$ C
  optimal operating temperature $\sim -110^\circ$ C
  lowest achievable temperature $\sim -134^\circ$ C
Format 2048 (cols) $\times$ 2049 (rows)
  24 $\mu$m square pixels
CTE 0.99998-0.99999
Dark Current negligible at $-110^\circ$ C
Full well $>150,000$ electrons/pixel
Readout maximum ADU = 65535
  4 available readout amplifiers (A,B,C,D)
  single, dual, or quad readout
  configurable subarray readout capability
Low Gain (C1S) Amp C, Gain Set 1.0, Slow Integrate, No MPP
  Bias 3989 ADU, Gain 6.1 $e^{-}\,ADU^{-1}$, Read Noise 4.5 $e^{-}$
High Gain (C2S) Amp C, Gain Set 2.0, Slow Integrate, No MPP
  Bias 1237 ADU, Gain 2.8 $e^{-}\,ADU^{-1}$, Read Noise 7.8 $e^{-}$

Fig. 1 is a plot of quantum efficiency vs. wavelength for the CCD in the GENII camera.

Figure 1: Plot of CCD chip quantum efficiency vs. wavelength. The relevant curve for our chip is the top solid curve.
\epsfig{file=gen2/,height=3.0in} \end{center}\end{figure}

The Dewar


The CCD is mounted on a cold finger in an evacuated chamber behind a fused silica window in a liquid nitrogen dewar and is cooled by direct contact of the cold finger with liquid nitrogen. The dewar, manufactured for ARC by Infrared Laboratories, Inc., has a capacity of 1 ${\ell}$ and a hold time of about 36 hours. The dewar is capable of cooling the CCD to a temperature as low as about $-134^{\circ}$C, but normally a heating resistor in the dewar is used to regulate the CCD temperature to some optimal working value between $-110^{\circ}$ and $-100^{\circ}$ C.

Normal Operation

  1. At least 24 hours before your scheduled observing night, send email to Nick Nichols ( and ask him to make sure the CCD dewar is filled. Top off the dewar at the beginning of the night, and again at the end of the night as a courtesy to the next observer if observations are scheduled for the following night.

  2. Filling the dewar takes about 15 minutes if it is still cold from the previous filling, but up to 40 minutes if starting from room temperature. It takes about 4 hours for the dewar to cool from room temperature to the optimal operating temperature of $-110^\circ$ C, so it is important to allow sufficient time for the cooldown before observing.

  3. To fill the dewar, attach the connector at the end of the hose from the dewar to a 25$\ell$ LN2 tank and fill the dewar with liquid nitrogen until you see liquid spilling out the side vent of the dewar connector. It takes some time for the hose to cool sufficiently to allow nitrogen to pass without evaporating, and the connectors will become coated with frost.

Potential Problems and the Dewar Vacuum

As of January 2007 the GenI and GenII CCD dewars have new vacuum valves which share a single new gauge which can be connected to either dewar. The vacuum is good for several days without pumping as long as the dewar is not allowed to warm up. The LN2 hold time is about 24 hours.

As a routine, keep filling the dewar every 24 hours or so as long as the camera is in use on the telescope. Leave the camera control software up and running on crux to check the temperature, with temperature regulation set for $-110^{\circ}$C. (For the GenI dewar be sure the defogging fan on the telescope tailpiece is running to keep frost from forming on the dewar window. This requires the CCD switch on the rack panel in the control room to be ON, to supply power to the camera controller and the defogging fan.)

The equipment for reading the vacuum gauge can usually be found in the storeroom on the dome floor level, or in the spectrograph room. It consists of a power supply transformer wired up to a 9-pin D connector and a digital multimeter. The D connector should be plugged into the connector on the vacuum gauge (before plugging in the power supply). When the meter is switched on to the 2VDC scale the voltage should ideally read 1.000V, which translates to roughly 0.01$\mu$ (0.01mTorr). Every increase of 1V is a factor of 10 in pressure, so 2V would be $\sim0.1\mu$, 3V is $\sim1\mu$, and 4V is $\sim10\mu$. According to the dewar manual, problems (such as outgassing and difficuly holding LN2) set in when the pressure reaches 5$\mu$, so for our purposes the vacuum is lost if the reading is over 4V. In practice, the dewar will probably not hold LN2 unless the reading is 2.5V or less.

To pump the dewar, connect the stainless steel hose from the vacuum pump to the dewar flange with a Quick Flange (QF) connector, but leave the dewar valve closed. The seal is made by compression of an O-ring between mating flanges by finger closure of a wingnut on a metal clamp, and the connectors on the dewar and the pump hose should be kept sealed with cover flanges when they are not connected to each other.

Plug in the vacuum pump to a 220 VAC outlet, using the extension cord if necessary. Press the PUMPING button to turn it on. The vacuum pump is a two-stage pump system which includes a controller. The roughing pump operates by itself first. The turbopump should spin up automatically when the roughing pump has lowered the pressure far enough for the turbopump to safely operate. Allow the turbopump to evacuate the hose for at least 30 minutes. After that time, if the turbopump is spinning (check the speed indicator if you can't hear it), the pressure should be low enough to safely open the vacuum valve on the dewar. (If the turbopump is not spinning, do not open the dewar vacuum valve! If you cannot find a leak in the hose or fittings that you can repair, the pump may need maintenance.) If the dewar pressure reading is not less than 3V ($\sim1\mu$) after 3 hours, there is probably a leak of some kind which must be fixed before the camera can be used.

When the pressure reading has dropped to 2.5V, close the dewar valve, turn off the vacuum pump, and fill the dewar. Ideally, the dewar will fill completely and the pressure reading will drop to 1.0V ($\sim0.01\mu$). If the dewar does not fill completely in less than 20 minutes, let it cool down for an hour or more, check to see that the pressure is still low and pump again if necessary, then try filling it again. As long as the dewar is kept filled and the pressure reading remains less than 2.5V the camera should work properly. When turning off the vacuum pump, wait until all rotor motion has stopped completely before unplugging the cord from the power outlet.

Operating the CCD Camera


  1. Log onto crux as user bench with password me4bench.

  2. Before proceeding, insert a blank tape into the DAT drive and enter the command mt -f /dev/rmt/0n status in any terminal window to verify that the tape drive is working. (Check the label on the tape drive for the device name currently in use.) You should get a message resembling:

    crux% mt -f /dev/rmt/0n status
    Sony 4mm DAT tape drive:
       sense key(0x6)= Unit Attention   residual= 0   retries= 0
       file no= 0   block no= 0

    If you don't get the above message try power cycling the tape drive. Saving your data to tape is one of the last and most important things you'll do at the end of the night, so it's best to make sure this will go smoothly at the outset.

  3. The home directory in which you will be working on crux is /crux/bench. In this directory are the directories fobos, which may be used to store setup files for the camera controller, and iraf, which contains the file, a startup file used by the IRAF program.

    In addition to the Sun internal 19GB disk, there is also an external 34GB disk attached to crux which appears as a directory called /data. All raw image files should be stored in the subdirectory /data/bench. In this directory (i.e. after entering cd /data/bench), create a unique subdirectory for your images.

Starting IRAF

First start the DS9 image display program by entering ds9 & at the prompt in a terminal window. Then open an xgterm terminal by entering xgterm &. In the xgterm window, from directory /crux/bench/iraf, enter cl to start IRAF. To see a help page for any IRAF task, enter help task at the cl> prompt. One way to run any IRAF task is to enter epar task, edit any parameters that you want to set or change, then type :go and hit RETURN. Tasks can also be run directly from the IRAF command line.

Starting the Voodoo Program

  1. Start the Voodoo camera control progam by selecting the Voodoo icon from the Gnome control panel. The Voodoo Main window (Fig. 2) should appear on the screen.

    Figure 2: The Voodoo Main window.
\epsfig{file=gen2/,width=4.0in} \end{center}\end{figure}

  2. Many configuration parameters for Voodoo may be set using the popup windows available from the menu bar of the Main window. First select Setup from the menu bar to bring up the Setup window (Fig. 3). Load the file /crux/bench/fobos/C1S.setup for Low Gain or the file /crux/bench/fobos/C2S.setup for High Gain, and click Apply to initialize the camera controller, then close the Setup window.

    Figure 3: The Voodoo Setup window.
\epsfig{file=gen2/,width=4.0in} \end{center}\end{figure}

  3. Select Parameters from the menu bar to bring up the Controller Parameters window. Select the Temperature tab, set the array temperature control to -110.0 C, and click Apply Above (Fig. 4). Select the Readout tab, choose Amplifier C, for example, and click Apply Above (Fig. 5). This currently seems to be the best readout amplifier for general purposes. Close the Controller Parameters window.

    Figure 4: The Voodoo Controller Parameters window, Temperature tab.
\epsfig{file=gen2/,width=4.0in} \end{center}\end{figure}

    Figure 5: The Voodoo Parameters window, Readout tab.
\epsfig{file=gen2/,width=4.0in} \end{center}\end{figure}

  4. Select Debug from the menu bar, then open the Developer Parameters window by selecting Development. Select the Gain tab and set, for example, Video Gain 1.0 and Integrator Speed Slow, then click Apply Above (Fig. 6). Close the Developer Parameters window.

    Figure 6: The Voodoo Developer Parameters window, Gain tab.
\epsfig{file=gen2/,width=4.0in} \end{center}\end{figure}

  5. Select Subarray from the menu bar to bring up the Subarray window and configure the subarray as desired. See Fig. 7 for an example. Click Apply to apply the settings. To revert to Full Array operation, select Full Array and click Apply. Close the Subarray window.

    Figure 7: The Voodoo Subarray window.
\epsfig{file=gen2/,width=4.0in} \end{center}\end{figure}

  6. If you would like your image headers to contain FITS keywords other than those required for basic formatting, you can use the FITS window (Fig. 8). If TCSLink is checked, the FITS header parameters labeled Universal Time, Local Sidereal Time, Equinox, Airmass, Hour Angle, Right Ascension, and Declination are updated automatically over a serial link to the telescope control PC at the beginning of each exposure or whenever you click Update. The FITS header parameters with white backgrounds may also be edited manually.

    Figure 8: The Voodoo FITS window.

Scope Control System

A Scope Control window (Fig. 9) has been added to Voodoo to allow the user to load an object list and command the telescope to slew to a selected list object using the serial link to the telescope control PC.

An object list must be a simple text file with extension *.lst, with one line per object. The format of each object line is arbitrary and may include any number of fields of any reasonable length, except that each line must include the RA and DEC separated by whitespace (spaces or tabs) only, each in sexagesimal format with no whitespace padding.

Figure 9: The Voodoo Scope Control window.
\epsfig{file=gen2/,width=4.0in} \end{center}\end{figure}

  1. In the Voodoo Scope Control window, click Load to select and load an object list file.

  2. Enter the Equinox of the coordinate list in the Equinox text field. This must be done only once per loaded list and may be changed at any time.

  3. To slew the telescope to a list object, swipe the coordinate section of the object line with the mouse (RA and DEC fields together) so that it is highlighted, then click Slew. The slew commmand will be echoed in the main Voodoo Information Window and the TCS will immediately slew the telescope. As always, a slew can be aborted with the Stop slew command (8) from the TCS Movement menu.

  4. Object lists cannot be edited or saved from the Scope Control window.

Taking an Exposure

  1. Check Save to Disk in the Main window and enter the full pathname of your image directory and a beginning filename for your images. If Auto Incr is checked, the numeric part of the filename will be automatically incremented with each new exposure. Otherwise you must enter a new filename for each new image.

  2. For normal exposures, check Open Shutter, enter the desired exposure time, and click Expose in the Main window.

  3. Display and analyze the images with IRAF and DS9.

Ending a CCD Session

  1. To copy your image files to DAT tape, change to your image directory in any teminal window, then use the unix command
    tar cvf /dev/rmt/0n .

    to write all your image files to tape. (This takes about half an hour for 100 images.)

  2. After your images have been written to tape, rewind the tape and take the tape drive off line by entering mt -f /dev/rmt/0 rewoffl, then remove your tape from the drive. Exit Voodoo, log out of IRAF, quit DS9, exit xgterm, and log out of the CDE window system.

next up previous contents
Next: The Fan Observatory Bench Up: manual Previous: The GenI CCD Camera

[an error occurred while processing this directive]